Authors: James Reagan (UMD/ESSIC/CICS-MD & NOAA/NCEI) and Tim Boyer (NOAA/NCEI)

Title: The World Ocean Database: Connecting Ocean Observing Systems since 1998

Type of Presentation: Oral

Short Abstract: The World Ocean Database (WOD) is the world’s largest in situ ocean profiling database

that is available without restriction. The WOD contains over 14.5 million quality controlled, uniformly

formatted profiles that date back to Captain Cook’s second voyage in 1772. The WOD merges data from

many different observing systems and instruments. There are three general steps that each dataset

received for inclusion in WOD goes through: the dataset is converted into a common format, automatic

and manual quality control checks are applied, and any incoming data that is already in WOD is removed

before merging (i.e., duplicate checking). The direct uses of WOD data are plentiful, ranging from

oceanographic and climate research to serving as environmental variables in biological studies. The

WOD also serves as the foundational dataset in calculating the World Ocean Atlas, a globally gridded

climatology of multiple oceanographic variables. Furthermore, WOD is also used to calculate ocean heat

and salt content products. While the WOD itself connects multiple ocean observing systems, the data

and products derived from WOD can also serve as validation datasets for various satellite missions,

ultimately connecting in situ observations to remotely sensed observations. The WOD and its utility in

both the in situ and remote sensing fields will be discussed.

The World Ocean Database: Connecting Ocean Observing

Systems since 1994

1University of Maryland – ESSIC/CICS-MD 2NOAA/NESDIS/NCEI Contact: james.reagan@noaa.gov

2016 NOAA CoRP Science Symposium July 18-19, 2016 Fort Collins, CO

James Reagan1,2, Tim Boyer2, Melissa Zweng2

Outline

• WOD Background • WOD Processing

– Conversion to standard format – Quality control checks – Duplicate checking

• WOD Uses – Develop climatological products – Climate and oceanographic research – Satellite calibration/validation

What is the World Ocean Database?

• A hydrographic database that contains over 14.8 million unique, quality controlled, and uniformly formatted oceanic profiles dating back to Captain James Cook’s second voyage in 1772.

Captain James Cook – His first measurements of subsurface temperature in the far Southern Ocean revealed water temperatures 100 fathoms (~183 meters) below the surface were warmer than water temperatures at the surface (Cook, 1777).

Portrait by Sir Nathaniel Dance-Holland (Wikipedia)

What instruments are in the World Ocean Database?

• Bottles (OSD)

• Conductivity, Temperature, and Depth (CTD) • Expendable Bathythermograph (XBT)

from NOAA Library

from NOAA Library

from NOAA Library

Instruments cont…

• Argo profiling floats (PFL)

• Gliders (GLD) • Autonomous pinniped bathythermograph

(APB)

from Boehlert et al. (2001)

from Scripps

from NOAA Library

Instruments cont… • Drifting Buoys (DRB)

• Moored Buoys (MRB)

• Surface Data (SUR) from NOAA/PMEL

from WHOI

from Sea-Bird

• Undulating Oceanographic Recorders (UOR)

• Mechanical Bathythermograph (MBT)

from NCEI

from WHOI

Temperature observing systems through the 20th and early 21st

Century 1934 : Nansen Cast 1960 : MBT

1985 : XBT 2009 : Argo

Red=Nansen Cast /CTD[1890s/1964], Light Blue=MBT [1939] Dark Blue=XBT [1967], Green=Argo float [2001] Orange=Tropical buoy [1984]

(Abraham et al., 2013)

Time Series of Data Additions to WOD by Instrument Type

(Chart courtesy of M. Zweng)

WOD Variables

• Multiple variables (e.g., temperature, salinity, oxygen, etc.) come from multiple instrument types.

• Blue indicates variables that undergo manual quality control in addition to automatic quality control checks (more on this later).

(Table courtesy of M. Zweng)

WOD Processing

Conversion to a Common Format

• Suite of Fortran tools to extract relevant data and metadata from files, convert to standard units, and store in the WOD binary format

• Standard formats make conversion easier! – Over the years, conversion

routines have been written for many different formats

– Best case, the submitter used a standard format so the converter can use existing code

– Worst case, they used a non-standard format and provided little metadata on that format

– NetCDF templates are nice!

(Slide courtesy of M. Zweng)

Quality Control Checks • Automatic QC Checks

• Standard deviation checks based on 5°x5° box statistics computed from all data in WOD

• Range checks • Density inversion checks • Cruise speed, significant figure, depth checks

• Manual QC Checks • “Bullseye” Checking

• Performed on gridded products (e.g., WOA13) produced from in situ data within WOD.

• Removal of offending profile causing the “bullseye”.

July 2011 Salinity Anomaly at 0m

Duplicate Checking • Self-Check

• Check to see if any incoming profiles are duplicates of another incoming profile

• Main Database Check • Check to see if any incoming

profiles are duplicates of profiles already in the main database (WOD)

• Multiple criteria are used • Value check(s) • Location/time check(s)

(Graphic courtesy of M. Zweng)

WOD Uses

(Graph courtesy of A. Grodsky)

WOD/WOA Citation Count by Year

WOD: Foundation for the Development of Global Ocean Climatologies and Anomalies

World Ocean Atlas Ocean Heat and Salt Content Anomalies

WOD: Foundation for Ocean Climate Research

Ocean Heat Content Sea Level Changes

WOD: Cal/Val of Satellite Data

Aquarius/SMOS Validation with WOD-derived Salinity

Updated from Reagan et al. 2014

September 2011 – May 2015

Conclusions

• WOD has successfully integrated multiple in situ ocean observing systems into one large database

• Three basic steps are followed during WOD processing 1. Data Conversion to Common Format 2. Data Quality Control 3. Data Duplicate Checking

• WOD has been utilized in the creation of multiple gridded oceanic products, ocean/climate research, and in the calibration and validation of satellite data.

Acknowledgements

Special thanks to Tim Boyer and Melissa Zweng for providing multiple figures/slides for this presentation. I would also like to thank the data providers who continue to send NCEI their data, my NCEI colleagues who continue to archive this data, and my Ocean Climate Laboratory colleagues who continue to process this data for inclusion in WOD.

Thank You